Lee-Carter模型在死亡率预测中的应用

Application of Lee-Carter Model in Predicting Mortality

摘要: 人口死亡率变化对社会保障系统有一定影响，它是时变的动态随机变量。基本养老保险的给付须遵从人口动态死亡率变化规律。本文利用Lee-Carter模型对未来人口死亡率变动趋势进行预测。对未来时间因子κ_t的预测时，用广义差分模型预测。并与ARIMA模型比较，结果广义差分模型预测效果较好。

Abstract: Changes in mortality rates have an impact on the social security system, which is a time-varying dynamic random variable. Basic pension insurance payment shall comply with the dynamic changes of population mortality. In this paper, using Lee-Carter model forecasts the future mortality trends of population, the prediction of time factor κt in the future, the paper uses generalized difference model for prediction. And compared with the ARIMA model, the result of generalized difference model was better.